← Back to community benchmarks
deepsweet-Qwen3.6-27BVL-oQ6
Performance
128k
tokens
417.7
PP tok/s
16.5
TG tok/s
313770
TTFT (ms)
43.6
Peak mem (GB)
Hardware
Chip
M5 Max (40c)
Memory
64 GB
GPU Cores
40
Software
oMLX
v0.4.4
macOS
macOS 26.5.1
Context
131,072
Performance by Context Length
| Context | PP tok/s | TG tok/s | Peak Mem | |
|---|---|---|---|---|
| 1k | 723.7 | 25.0 | 22.4 GB | view |
| 4k | 851.6 | 24.0 | 23.8 GB | view |
| 8k | 799.5 | 24.4 | 24.5 GB | view |
| 16k | 731.4 | 23.6 | 25.7 GB | view |
| 32k | 642.7 | 22.1 | 28.2 GB | view |
| 64k | 558.9 | 19.3 | 33.2 GB | view |
| 128k | 417.7 | 16.5 | 43.6 GB | current |
Batching Results
| Batch Size | TG tok/s | Speedup |
|---|---|---|
| 1× | 25.0 | 1.00× |
| 2× | 37.0 | 1.48× |
| 4× | 93.1 | 3.72× |